⚡ 4 h early
Anthropic investigates cybersecurity evaluation incidents
Investigating three real-world incidents in our cybersecurity evaluations
13 experts across 5 network communities independently surfaced this.
Why this matches
Concern & critique reaction
3 attributable expert contributions
· Simon Willison, René Walter, Andrea Lathrop
“Anthropic had previously attacked PyPI, but this OpenAI attack on RubyGems was a whole lot more aggressive https://t.co/8ez7MnqxTw https://t.co/npI3RHwirQ” evidence ↗
13 experts
5 communities
1 sources clustered
Mark Riedl
Simon Willison
Grace
Dare Obasanjo
Vincent Conitzer
Liz Fong-Jones (方禮真)
Elie
René Walter
+5
How the network is reacting
The reaction spans practical upside and material concerns.
Live tension
Markets & investment
4 expertsCapital, companies and commercial impact.
“Claude hacked 3 systems thinking it was part of simulated evaluations www.anthropic.com/news/investi...”
Concern & critique
3 expertsRisks, limits and unintended consequences.
“Anthropic had previously attacked PyPI, but this OpenAI attack on RubyGems was a whole lot more aggressive https://t.co/8ez7MnqxTw https://t.co/npI3RHwirQ”
2 experts discussed this · 4 posts
Tim Duffy: Anthropic announces they've also had models gain unauthorized access during evaluations www.anthropic.com/news/investi...
Tim Duffy: Compared to the OpenAI one these are maybe less evidence of misalignment, since the models were wrongly given internet access.
Grace: Anthropic announces they've also had models gain unauthorized access during evaluations www.anthropic.com/news/investi...
Open the full discussion →